E-ring-modified 7-oxyiminomethyl camptothecins: Synthesis and preliminary in vitro and in vivo biological evaluation

Article abstract of DOI:10.1016/j.bmcl.2008.03.074

In contrast to five-membered E-ring analogues, 7-oxyiminomethyl derivatives of homocamptothecins showed ability to form stable ternary complexes with DNA and topoisomerase I. The 7-oxyiminomethyl derivatives of homocamptothecins were evaluated as a racemi

Full text of DOI:10.1016/j.bmcl.2008.03.074

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2910–2915
E-ring-modified 7-oxyiminomethyl camptothecins: Synthesis
and preliminary in vitro and in vivo biological evaluation
Giuseppe Giannini,a,* Mauro Marzi,a Walter Cabri,a Elena Marastoni,a
Gianfranco Battistuzzi,a Loredana Vesci,a Claudio Pisano,a Giovanni Luca Beretta,b
Michelandrea De Cesareb and Franco Zuninob,*
aSigma-Tau Research & Development, Via Pontina Km 30.400, I-00040 Pomezia, Rome, Italy
bFondazione IRCCS Istituto Nazionale per lo Studio e la Cura dei Tumori, Via Venezian 1, 20133 Milan, Italy
Received 15 February 2008; revised 26 March 2008; accepted 27 March 2008
Available online 30 March 2008
Abstract—In contrast to five-membered E-ring analogues, 7-oxyiminomethyl derivatives of homocamptothecins showed ability to
form stable ternary complexes with DNA and topoisomerase I. The 7-oxyiminomethyl derivatives of homocamptothecins were eval-
uated as a racemic mixture. Following the isolation of the two enantiomers, the 20 (R)-hydroxy isomer confirms the best activity. By
using a panel of human tumor cells, all tested homocamptothecins showed a potent antiproliferative activity, correlating to the per-
sistence of the cleavable complex. No significant difference was observed between the natural scaffold and the corresponding homo-
camptothecin homologue. A selected compound of this series exhibited an excellent antitumor activity against human
gastrointestinal tumor xenografts.
Ó 2008 Elsevier Ltd. All rights reserved.
Camptothecin and its analogues exhibit a broad spec-
trum of antitumor activity, representing a very promis-
ing class of anticancer agents currently used in clinical
practice and characterized by a selective activity as topo-
isomerase I inhibitor.1
Chemically, camptothecins are characterized by a penta-
cyclic moiety with an a-hydroxy lactone ring (E-ring,
Fig. 1) which, under alkaline conditions, is present in
its less potent open form inhibiting the topoisomerase
I enzyme and in vivo potency. Modifications in the E-
ring generally reduce or abolish this activity. In 1997,
Lavergne et al., aiming to improve the drug’s stability
in human plasma, synthesized a novel seven-membered
E-ring homologue, where the natural a-hydroxy-d-lac-
tone E-ring was replaced by a b-hydroxy-e-lactone ring.
These new derivatives, called homocamptothecins
Figure 1. Structures of corresponding CPT and hCPT homologues
(hCPT),2 showed an activity comparable to that of the
and Structure of SN38.
natural scaffold (Fig. 1).
In this decade, a number of hCPT derivatives were syn-
thesized not only by the total synthesis, but also by the
following racemic and enantioselective synthetic strate-
gies, starting from the natural scaffold. Some of these
hCPTs, such as diflomotecan (currently in Phase II
clinical trials),3–5 were evaluated in many clinical trials.
Since the introduction of a methylene group in the
Keywords: Topoisomerase
Antitumor.
I
inhibitors; Homocamptothecins;
*
Corresponding authors. Tel.: +39 0691393640; fax: +39 0691393638
(G.G.); tel.: +39 02 23902267 (F.Z.); e-mail addresses: giuseppe.
giannini@sigma-tau.it; franco.zunino@istitutotumori.mi.it
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.03.074

G. Giannini et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2910–2915
2911
R
N
the corresponding 7-hydroxymethyl derivative; this
intermediate was converted in situ to 7-dimethylacetal
derivative 4 by oxidation with MnO2 in MeOH. NaBH4
was added to a solution of 4 in MeOH. A subsequent
reaction with NaIO4 in CH3COOH gave intermediate
5. Under Reformatsky conditions 5, with tert-butyl bro-
moacetate, gave b-hydroxyester 6 which, by treatment
with O-tert-butyl hydroxylamine hydrochloride or O-
benzyl hydroxylamine hydrochloride in CH3COOH, at
80 °C, was directly converted into 2a and 3a,
respectively.
8. R = H
R' = CH2COOtBu
R' = CH2COOH
O
9. R = H
N
O
10. R = CH=NO
t
Bu
R' = CH2COOH
R' = OH
O
11. R = CH=NOtBu
R'
12. R = CH=NOCH2Ph
R' = CH2COOH
Figure 2. Structures of E-ring-modified CPTs [8: ST2084; 9: ST2085;
10: ST2196; 11: ST2715; 12: ST2285].
E-ring of CPT provides enhanced stability and de-
creased protein binding in human plasma without
affecting its capability to poison topoisomerase I-
DNA complex, the present study was undertaken to
explore whether this modification improves the biolog-
ical properties of the 7-oxyiminomethyl derivatives6,7
(Fig. 1). The most advanced derivative of this series
is gimatecan, currently in Phase II clinical trials.8 This
is the first study on 7-oxyiminomethyl-CPT modified in
the E-ring.
The racemic mixture of 2a was separated by prep HPLC
(Scheme 2) into the two single enantiomers (2b and 2c),
using a chiral stationary phase.
According to what was expected with the hydroxyl
group in specific C-20 configuration, 20(R)-hydroxyl-
hGMT (2b) was more active than 20(S)-hydroxyl-
hGMT (2c).
Another series with a modified E-ring was also prepared
(8–12; Fig. 2). According to the previous observations,9
the five-membered E-ring analogues exhibited a mark-
edly reduced cytotoxic activity (IC50 ꢀ 1 lM, 2-h expo-
sure against H460 cells).
The 7-oxyiminomethyl hCPT derivatives (Scheme 1)
were prepared in enantiomeric mixture by the synthetic
approach cited above (Lavergne).2 CPT was treated
with sulfuric acid, FeSO4Æ7H2O and 40% H2O2 to give
OMe
MeO
OMe
MeO
a
O
b
O
CPT
N
N
N
O
N
O
O
4
5
H
O
O
OH
c
R
O
N
OMe
MeO
O
*
d
O
N
N
OH
N
N
O
O
2a-3a
*
6
O
HO
O
HO
Scheme 1. Synthesis of 7-oxyiminomethyl hCPT analogues 2a and 3a. Reagents and conditions: (a) 96% H2SO4, FeSO4Æ7H2O, 40% H2O2, MnO2,
MeOH, 10–50 °C (83%). (b) i—NaBH4, MeOH, rt (78%); ii—NaIO4, CH3COOH, rt (89%). (c) Zn, anhydrous Et2O (CH3)3SiCl, BrCH2COOtBu,
anhydrous THF, reflux (57%). (d) RONH2ÆHCl, CH3COOH, 80 °C (2a R = ꢁt-Bu 56%, 3a R = ꢁCH2Ph 45%).
O
N
O
N
O
O
PrepHPLC
N
N
2a
+
N
N
O
O
O
O
HO
HO
2c: ST2523
2b: ST2522
Scheme 2. Separation of the two enantiomers 2b and 2c. [Column (S,S)-DACH-DNB 5/100, eluent: CH2Cl2/n-hexane/MeOH (80:20:0.1); 1 mL/min,
22 °C, k = 360 nm].

2912
G. Giannini et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2910–2915
R
R
a
O
b
O
N
CPT
N
N
O
N
O
O
4 R=CH(OCH3)2
5 R=CH(OCH3)2
5' R=H
H
O
O
OH
c
R
R
O
d
O
*
N
N
O
OH
N
N
O
O
O
7 R=CH(OCH3)2
8 R=H
6 R=CH(OCH3)2
6' R=H
O
O
HO
e
f
R'
O
N
O
N
O
O
N
N
O
O
N
O
9
O
O
OH
10 R'=tBu
12 R'=CH2Ph
OH
Scheme 3. Synthesis of five-membered E-ring CPT analogues 8–9–10–12. Reagents and conditions: (a) 96% H2SO4, FeSO4Æ7H2O, 40% H2O2, MnO2,
MeOH, 10 °C to 50 °C (4: 83%). (b) i—NaBH4, MeOH, rt (R = H 75%, R = CH(OCH3)2 78%); ii—NaIO4, CH3COOH, rt (R = H (50) 80%,
R = CH(OCH3)2 (5) 89%). (c) Zn, anhydrous Et2O (CH3)3SiCl, BrCH2COOtBu, anhydrous THF, reflux (6: 58%; 60: 60%). (d) PDC, CH2Cl2 rt (7:
74%; 8: 73%). (e) R0ONH2ÆHCl, CH3COOH, 80 °C (10: 68%, 12: 61%). (f) TFA, CH2Cl2 rt (9: 100%).
An analogue approach was used to obtain intermediates
6–60 in the synthesis of five-membered E-ring analogues
8–9–10–12 (Scheme 3). b-Hydroxyester 6–60 (with
R = CH(OCH3)2 or H), by treatment with PDC (Pyrid-
O
O
N
N
O
a
N
O
N
inium DiChromate) in CH2Cl2 at rt, gave the five-mem-
bered E-ring intermediates 7 and 8; this latter, under
hydrolysis with TFA in CH2Cl2, gave 9 in quantitative
yield. The subsequent reaction of 7 with O-tert-butyl
hydroxylamine hydrochloride or O-benzyl hydroxyl-
amine hydrochloride in CH3COOH at 80 °C led to10
and 12, respectively.
N
O
N
O
O
2
H
13
O
O
OH
b
O
Compound 2 was converted into b-hydroxyester (13) by
a two-step reaction (Scheme 4), NaBH4 in MeOH and
subsequently NaIO4 in CH3COOH. A cyclization reac-
tion with K2CO3 in MeOH/H2O (10:1 v/v) gave 11.
N
O
N
N
O
The antiproliferative effects of the hCPTs were deter-
mined after 72 h following a 1-h exposure to the drugs.
Compared to topotecan [9-dimethylaminomethyl-10-hy-
droxy-camptothecin] and SN38 [7-ethyl-10-hydroxy-
camptothecin], the hCPTs studied (Table 1) exhibited
a higher potency in inhibiting cell growth.
11
HO
Scheme 4. Synthesis of 11. Reagents and conditions: (a) i—NaBH4,
MeOH, rt (78%); ii—NaIO4, CH3COOH, rt (89%). (b) K2CO3,
MeOH/H2O (10:1 v/v) (65%).
As expected, the expansion of the lactone ring of CPT
produced an increased antiproliferative activity (1 vs
1a). The cytotoxic potency of 2a and 3a was comparable
to that of the six-membered CPT, 2 and 3, respectively.
The expansion of the lactone ring of gimatecan was
slightly detrimental for cytotoxic potency (2 vs 2a) while
an increased activity was observed for benzyl derivative
(3 vs 3a). In addition, compared to 2a, compound 3a
exhibited an increased cytotoxic potency. These features
indicated the advantage to have a lipophilic substituent

G. Giannini et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2910–2915
Table 1. Antiproliferative activity of 7-modified homocamptothecins
2913
a
Drug
IC50 (lM)
H-460
1.18 0.24
HT29
HT29/mit
RI
Topotecan
SN38
1
1.35 0.66
0.67 0.36
3.16 1.2
46.29 21
11.5 2.7
34
17
0.22 0.013
0.18 0.01
8.81 2.35
8.55 2.48
0.070 0.0022
0.56 0.0867
0.042 0.014
0.35 0.022
2.8
19
1.2
14
0.75
20
1a
2
0.0496 0.047
0.02 0.005
0.0368 0.003
0.03 0.002
0.015 0.0003
0.015 0.006c
0.0018 0.0002c
>0.2c
0.44 0.0002
0.056 0.0223
0.041 0.0086
0.056 0.022
0.017 0.0021
2a
3
3a
2
2b
2c
a IC50, drug concentration required for 50% reduction of cell growth as compared to untreated controls after 1-h exposure to the drug. Means SD
are reported from at least three experiments.
bRI, resistance index in HT29/mit.
c IC50, determined after 2-h exposure.
in position 7 which could account for a more favorable
intracellular accumulation. During two hours exposure,
T C
M
20 (R) homogimatecan revealed a 10-fold increase in
cytotoxic activity in comparison to gimatecan (2 vs
2b). Although effective in inhibiting the proliferation
of HT29/mit, all compounds exhibited cross-resistance
in this subline which overexpresses BCRP (Breast Can-
cer Resistance Protein), a transport system implicated
in the resistance to conventional CPTs.10 The results
shown in Table 1 support that all tested homocamptot-
hecins (1a, 2a, and 3a) are better substrates for BCRP
than the corresponding CPT analogues with the natural
E-ring.
Topoisomerase I-mediated DNA cleavage assays with
purified human topoisomerase I were used to investigate
the ability of hCPT to stimulate DNA damage. The
hCPTs revealed an intensity of DNA damage compara-
ble to that of SN38, used as reference compound
(Fig. 3).
The cleavage pattern revealed additional sites not de-
tected in the presence of SN38 (arrows in figure). Since
the drug interaction with the DNA–enzyme complex is
expected to be reversible, the persistence of the cleavable
complex was evaluated after the addition of high salt
concentrations (0.6 M NaCl), which favors the dissocia-
tion of the ternary drug–enzyme–DNA complex. As
compared to SN38, 3a and 2a revealed a more stable ter-
nary complex. Such stability was particularly evident for
2a, which exhibited a DNA damage persistence of
around 90% after 10 min (Fig. 4). From the inhibition
data of topoisomerase I of the three compounds, 1a,
2a, and 3a, it is evident that the presence of a bulky
group in 7-position on B-ring of CPT scaffold, favors
tertiary complex stability, as demonstrated by the intro-
duction of a residual benzyloxyiminomethyl-group (3a)
and tert-butyloxyiminomethyl (2a) in this position.
Figure 3. Topoisomerase I-mediated DNA cleavage assays. The
samples were reacted with 1, 10, and 50 lM drug at 37 °C for
30 min. Reaction was then stopped by adding 1% SDS, 0.3 mg/mL of
proteinase K, and incubating for 45 min at 42 °C before loading onto a
denaturing 8% polyacrylamide gel. C, control DNA; T, reaction
without drug; M, purine markers. Arrows indicate the additional sites
observed for hCPTs. The experiment was repeated three times,
reporting the results of a representative value.
panel of human tumor xenografts. The maximum toler-
ated dose (MTD) was determined with an intermittent
q4dx4 schedule. In animals bearing MKN-28 gastric
carcinoma, 2a was well tolerated up to 12 mg/kg (Table
2). Derivative 3a was substantially less effective under a
comparable range of doses. Therefore, 2a was also
tested against CoBA colon carcinoma and pancreatic
carcinoma, Panc-1. Against CoBa, a dose of 15 mg/kg
was effective without lethal toxicity, but with substantial
body weight loss (Table 2 and Fig. 5). A 20 mg/kg dose
was highly toxic in most treated mice (3/4). With a dose
of 14 mg/kg against the pancreatic carcinoma model
On the basis of their antiproliferative potency, 3a and 2a
were selected for further preclinical development (Table
2). The antitumor activity of oral 2a was studied in a

2914
G. Giannini et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2910–2915
100
80
60
40
20
0
♦
♦
♦
1000
SN38
2a
3a
100
♦
1a
0
2
4
6
8
10
Time (min)
10
0
15
30
45
60
75
Figure 4. Cleavage persistence assay. The samples were reacted for
30 min with 10 lM drug. DNA cleavage was then reversed by adding
0.6 M NaCl. The 100% value refers to the extent of DNA cleavage
after 30 min of incubation. Each value was obtained by densitometric
analysis. The experiment was repeated three times, reporting the results
of a representative value.
Days after tumor implant
Figure 5. In vivo antitumor activity of 2a. (ST2127) on sc-growing
CoBa colon carcinoma. (ꢂ) Untreated control tumors; (m) oral 2a,
15 mg/kg, q4dx4. (h) CPT11, 50 mg/kg, q4dx4. Bars, standard
deviation.
Panc-1, there was excellent antitumor activity with no
evidence of lethal toxicity. Indeed, 9/10 treated tumors
exhibited complete tumor regression (Table 2). The
activity of 2a was superior to that of irinotecan at max-
imum tolerated doses under similar treatment
conditions.
enantiomers (2b and 2c) and the 20(R) isomer shows
more in vitro activity than the corresponding 20(S).
In addition, compared to GMT (2), 20(R) hGMT (2b),
reveals about a 10-fold increase in cytotoxic potency.
However, in contrast to GMT, hCPTs appear to be the
substrates of BCRP, which is expected, since it was
already reported that the expansion of the lactone ring
increase the recognition of the drug by BCRP.11 As
already reported for other compounds of this series
(e.g., GMT), the tested hCPTs exhibit stable ternary
complexes with DNA and topoisomerase I. On the basis
of the antiproliferative data reported in Table 1, and of
stabilization of the cleavable complex (Fig. 4), it is evi-
dent that the inhibition of the intracellular target is the
crucial event responsible for cytotoxic potency. Indeed,
the two potent hCPTs (2a and 3a) exhibited an increased
capability to stabilize the cleavable complex. Although
the stabilization of the cleavable complex is recognized
Among the many strategies adopted in the last years
aimed at improving lactone stability of CPT, the synthe-
sis of five- and seven-membered E-ring CPT appears to
be the most promising.12,3–5 The results reported in the
present manuscript indicated that the elimination of a
methylene group of the lactone ring is detrimental for
the activity of the studied CPTs. In contrast to five-mem-
bered E-ring analogues, 7-oxyiminomethyl derivatives of
hCPT exhibit a cytotoxic activity comparable to that of
the corresponding CPT with the natural lactone ring.
The racemic hGMT (2a) separates into the two single
Table 2. Effects of 3a (ST2143) and 2a (ST2127) given orally, q4dx4, against human gastrointestinal tumors xenografted sc in athymic nude mice
Tumor model
Drug
Dose (mg/kg)
TVI%a
CRb
LCKc
BWL%d
Toxe
MKN28
Gastric carcinoma
3a
3a
3a
2a
2a
2a
4
8
17
60
47
71
47
81
0/8
0/8
0/8
0/8
0/8
0/8
0
0
6
0
0
0
2
0/4
0/4
0/4
0/4
0/4
0/4
0.4
0.2
0.6
0.2
0.6
16
3
6
12
Panc-1
Pancreatic carcinoma
2a
2a
14
15
99
99
99
9/10
7/8
1.4
1.6
1.6
5
16
7
0/5
1/5
0/5
CPT11
400
5/10
CoBa
Colon carcinoma
2a
2a
15
20
2
95f
—
90
86
0/8
—
1.3
—
21
18
18
2
0/4
3/4
0/4
0/4
Gimatecan
CPT11 (iv)
0/8
0/8
1.1
1.1
50
a Tumor volume inhibition % in treated control mice.
b Complete responses: that is, disappearance of tumor lasting at least ten days.
c Gross log10 cell kill to reach an established tumor volume; the highest change is reported.
d Body weight loss % induced by drug treatment.
e Dead/treated mice.
f P < 0.05 vs CPT11, by Student’s t test.

G. Giannini et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2910–2915
2915
to be one of the most important feature for CPT activity,
References and notes
differences in cellular uptake and subcellular distribution
may contribute to determine antitumor activity. Such
observations suggest that optimal drug efficacy can be
achieved from a balance between cellular pharmacoki-
netics and intrinsic ability to induce DNA cleavage.
1. Rothenberg, M. L. Ann. Oncol. 1997, 8, 837.
2. Lavergne, O.; Lesueur-Ginot, L.; Pla Rodas, F. J.; Bigg,
D. C. H. Bioorg. Med. Chem. Lett. 1997, 7, 2235.
3. Russell, W. D.; Li-Xi, Y. Mini Rev. Med. Chem. 2005, 5,
425.
4. Bailly, C. Crit. Rev. Oncol. Hematol. 2003, 45, 91.
5. Scott, L.; Soepenberg, O.; Verweij, J.; de Jonge, M. J. A.;
Planting, A. S.; McGovern, D.; Principe, P.; Obach, R.;
Twelves, C. Ann. Oncol. 2007, 18, 569.
6. Dallavalle, S.; Ferrari, A.; Biasotti, B.; Merlini, L.; Penco,
S.; Gallo, G.; Marzi, M.; Tinti, M. O.; Martinelli, R.;
Pisano, C.; Carminati, P.; Carenini, N.; Beretta, G.;
Perego, P.; De Cesare, M.; Pratesi, G.; Zunino, F. J. Med.
Chem. 2001, 44, 3264.
A selected compound of this series (2a) produces
remarkable antitumor activity against human gastroin-
testinal tumors. The antitumor potency and efficacy of
the novel hGMT were apparently superior to that of
CPT11. The advantages of 2a over GMT remain to be
documented.
7. Perego, P.; De Cesare, M.; De Isabella, P.; Carenini, N.;
Beggiolin, G.; Pezzoni, G.; Palumbo, M.; Tartaglia, L.;
Pratesi, G.; Pisano, C.; Carminati, P.; Scheffer, G. L.;
Zunino, F. Cancer Res. 2001, 61, 6034.
8. Biron, P.; Reichardt, P.; Grosso, F.; Le Cesne, A.; Poveda,
A.; McCanna, S.; Zanna, C.; Brienza, S.; Ray-Coquard, I.;
Blay, J. Y. J. Clin. Oncol. 2007; ASCO Annual Meeting
Proceedings Part I, 25, 10063.
Acknowledgments
This work was partially supported by the Associazione
Italiana Ricerca sul Cancro, by the Ministero della Sa-
lute, and by the Fondazione Italo Monzino.
9. Hertzberg, R. P.; Caranfa, M. J.; Holden, K. G.; Jakas, D.
R.; Gallagher, G.; Mattern, M. R.; Mong, S. M.; Bartus,
J. O.; Johnson, R. K.; Kingsbury, W. D. J. Med. Chem.
1989, 32, 715.
10. Beretta, G. L.; Perego, P.; Zunino, F. Curr. Med. Chem.
2006, 13, 3291.
Supplementary data
Materials and methods: Growth Inhibition Assay; Topo-
isomerase I–Dependent DNA Cleavage Assay; Antitu-
mor activity studies. References for material and
methods. NMR spectra: ST2084; ST2085; ST2086;
ST2127; ST2143; ST2196; ST2285; ST2715. Polarimetric
Analysis of the two single enantiomers ST2522; ST2523.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.2008.
03.074.
´
11. Bates, S. E.; Medina-Perez, W. Y.; Kohlhagen, G.;
Antony, S.; Nadjem, T.; Robey, R. W.; Pommier, Y.
J. Pharmacol. Exp. Ther. 2004, 310, 836.
12. Takagi, K.; Dexheimer, T. S.; Redon, C.; Sordet, O.;
´
Agama, K.; Lavielle, G.; Pierre, A.; Bates, S. E.; Pommier,
Y. Mol. Cancer Ther. 2007, 6, 3229.